Currently we have a limited understanding of what happens from the moment editors receive the initial machine translation (MT) output of an article to when they publish edited versions of these same articles. To identify ways of supporting editors in their goal of high quality published translations, we foremost need to better understand the linguistic changes made during the MT post-editing process and how these differ by language, language type, and article characteristics. This project aims to provide an analysis of such changes by involving language experts to analyze actual MT outputs and CX-created articles. For more info about current active Language Team design research projects, please refer to https://airtable.com/shrxcU0kgFdEA4Ovb

Results: https://meta.wikimedia.org/wiki/Research:Machine_Translation,_Human_Editors